Natural products have historically been the source of most of the anti-mitotic small molecules whose properties have allowed them to become useful drugs. That remains true of most but not all of the compounds in this study. Some, such as the new microtubule-stabilizing compound peloruside, are natural products. Others, such as analogs of the microtubule-stabilizing compound epothilone, and analogs of the microtubule-destabilizing peptide tubulysin, are derived by synthesis based on the structure of known natural compounds. We have also discovered anti-mitotic compounds from a high-throughput screen of libraries of synthetic molecules whose structures are unrelated to particular natural compounds. All of these compounds exert their actions through binding to tubulin, the subunit protein of microtubules. We have also demonstrated that compounds like nitrosoureas, that do not interact with tubulin, can affect microtubule properties by interacting with small proteins that themselves bind to tubulin. We have identified a number of new modified peptides derived from the natural peptide microtubule destabilizer, tubulysin. These new peptides show a range of antimicrotubule activity in assays with purified proteins as well as in cells. We have also designed, synthesized, and tested a new, potent microtubule stabilizer based on the natural compound, epothilone. In addition we have examined a new "second generation" of microtubule stabilizers, focusing on the natural compound peloruside. Study of its binding to tubulin by hydrogen-deuterium exchange mass spectrometric methods has led to new insights into the mechansims of normal microtubule assembly. We also have developed new methods of analysis of mass spectrometric data for studying tubulin from tissues samples. Binding of small molecules to tubulin can be usefully followed by fluorescence correlation spectroscopy, as we have shown. We have also shown that an old chemotherapy agent, a nitrosourea, may affect microtubule stability indirectly by altering the activity of the protein stathmin. This results in reduced migration and invasion by malignant glioma cells. In addition to studying small molecules that may have useful activity against human tubulin, we are seeking to identify small molecules that do not bind well with mammalian tubulin but do bind to parasite tubulin. The tubulin molecule is quite conserved evolutionarily, but differences do exist, and several molecules are known that can target, for example, yeast rather than mammalian tubulin or vice-versa. We are looking for molecules that will target Leishmania, the infectious cause of an important group of human diseases. We have identified several small molecules that show promise as selective agents, binding to Leishmania tubulin preferentially over mammalian tubulin, and preventing parasite multiplication inside human macrophage cells.